TWI516039B - Flexible and scalable enhanced transmission selection method for network fabrics - Google Patents

Description

Transmission selection method for flexibility and scalability of network Fibre Channel networks

The present invention relates generally to assigning bandwidth in a network Fibre Channel network or, more specifically, to assigning a bandwidth to a particular logical data stream within a network Fibre Channel network.

Computer systems often use multiple computers that are coupled together in a common chassis. The computers may be separate servers coupled by a common backbone within the chassis. Each server is an insertable board that includes at least one processor, an onboard memory, and an input/output (I/O) interface. In addition, the servers can be connected to the switch to extend the capabilities of the server. For example, the switch can grant the server access to additional Ethernet or PCIe slots, as well as permit communication between servers in the same or different chassis.

A plurality of switches can be combined to form a logic switch. In addition, each physical connection between the plurality of switches and the server can be managed to assign a minimum amount of bandwidth to a particular data stream in the connection. That is, one of the physical connections may be assigned more bandwidth than the other. Increasing the ability of system administrators to assign bandwidth to specific streams also increases the system administrator's control over the network.

Embodiments of the present invention disclose a method and a computer readable storage medium for providing a physical connection between two computing devices, wherein network traffic flowing through the entity is logically divided into a plurality of virtual area networks (VLANs) aisle. The method and the computer readable storage medium are in the plurality of VLAN channels A bandwidth of the physical connection is allocated between at least two of the two, and a bandwidth allocated to one of the plurality of VLAN channels is further divided among the plurality of traffic classes.

In another embodiment, a network device includes a port configured to connect to a computing device via a physical connection, wherein network traffic flowing through the physical connection is logically divided into a plurality of virtual Local area network (VLAN) channel. The network device also includes a bandwidth scheduler configured to allocate one of the bandwidths between at least two of the plurality of VLAN channels, and in the plurality of traffic The classes are further divided into bandwidths assigned to one of the plurality of VLAN channels.

The embodiments of the present invention briefly summarized above will be described more specifically by reference to the accompanying drawings.

It is to be understood, however, that the appended claims

IEEE 802.1Q (or 802.1p) permits the logical division of data streams on a physical connection into eight different priority or traffic classes. The standard also provides techniques for splitting the total bandwidth of physical connections between eight traffic classes. For example, an administrator can set a minimum bandwidth that must be provided to a given traffic category. To provide additional flexibility, IEEE 802.1Qaz (hereafter referred to as Enhanced Transmission Selection (ETS)) provides a standard for reallocating bandwidth between traffic classes. When the network traffic assigned to a traffic class does not use its allocated bandwidth, ETS allows other traffic classes to use the available frequency. width. ETS can even coexist with strict priorities - that is, the bandwidth used for a particular traffic class cannot be shared. If the bandwidth used for the two traffic classes is not a strict priority, the ETS allows the unused bandwidth of one traffic class to be used by another traffic class. For example, if the data stream associated with two traffic classes is a cluster (ie, experiencing a high load for a short period of time), as long as the high loads do not occur simultaneously, the traffic category currently experiencing high load is Some or all of the bandwidths assigned to another traffic class may be used. If the high load occurs simultaneously such that each traffic class uses all of the allocated bandwidth, the ETS does not allow one traffic class to borrow bandwidth from the other.

IEEE 802.1Q also standardizes the use of virtual local area networks (VLANs) in a Converged Enhanced Ethernet (CEE) network. In general, VLANs permit end stations of a physical LAN to be grouped together, even if the terminal stations of the LAN are not located on the same network switch. That is, two physical connections associated with two different switches can be logically connected to be part of the same LAN (ie, VLAN). The packet includes a VLAN ID (i.e., a VLAN tag) that assigns the packets to a particular VLAN. VLANs provide segmentation services that are traditionally provided by routers and increase LAN scalability, security, and network management.

IEEE 802.1Q provides 12 bits for identifying Service VLANs (S-VLANs) - that is, 4096 uniquely addressable S-VLANs. Later routing standards add each S-VLAN to have a plurality of associated guest VLANs (C-VLANs). In IEEE 802.1QinQ, for example, each S-VLAN can be associated with up to 4096 C-VLANs. LANs that follow this standard can have up to 16,777,216 VLANs (2^12*2^12). however, IEEE 802.1Q and ETS provide only eight different traffic classes that can be used to control the bandwidth in a particular physical connection (or link). Instead of relying solely on these eight traffic classes to manage bandwidth, the embodiments discussed herein disclose the use of an enhanced ETS scheduler to assign the bandwidth of a physical connection to an individual VLAN. Instead of relying solely on the traffic class, the VLAN ID is used to permit the network device to allocate bandwidth to, for example, millions of unique VLANs. Therefore, the use of VLAN IDs can increase the granular control of network Fibre Channel networks and their performance.

In the following, reference is made to embodiments of the invention. However, it should be understood that the invention is not limited to the particular described embodiments. In fact, any combination of the following features and elements (whether or not related to different embodiments) is contemplated to implement and practice the invention. In addition, although an embodiment of the present invention may achieve advantages over other possible solutions and/or advantages over the prior art, whether a particular advantage is achieved by a given embodiment is not a limitation of the invention. Therefore, the following aspects, features, embodiments and advantages are merely illustrative and are not to be considered as an element or limitation of the scope of the appended claims. Also, the reference to "the invention" is not to be construed as a summary of any inventive subject matter disclosed herein, and The elements or limitations of the scope.

As will be appreciated by those skilled in the art, aspects of the invention may be embodied in a system, method or computer program product. Accordingly, aspects of the present invention may take the form of a complete hardware embodiment, a fully software embodiment (including firmware, resident software, microcode, etc.) or a combination of soft and hard aspects, such embodiments This article can be referred to as "circuit", "module" or "system" System." Additionally, aspects of the invention may be in the form of a computer program product embodied in one or more computer readable media having computer readable code embodied thereon.

Any combination of one or more computer readable media may be utilized. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. For example, a computer readable storage medium can be, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing systems, devices, or devices. More specific examples (non-exhaustive list) of computer readable storage media will include the following: electrical connections with one or more wires, portable computer magnetic disks, hard disks, random access memory (RAM), read only Memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable optical disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or the foregoing Any suitable combination of each. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

The computer readable signal medium can include a propagated data signal having a computer readable code embodied therein (eg, in a baseband or as part of a carrier). The propagated signal can take any of a variety of forms including, but not limited to, electromagnetic, optical, or any suitable combination thereof. The computer readable signal medium can be any computer readable computer that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with the instruction execution system, apparatus, or device. media.

Any suitable medium may be used to transmit the code embodied on a computer readable medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer code for performing aspects of the present invention may be written in any combination of one or more programming languages, including object oriented programs such as Java, Smalltalk, C++, or the like. A design language and a conventional procedural programming language such as a "C" programming language or a similar programming language. The code can be executed entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on the remote computer, or completely remotely. Executed on a computer or server. In situations where it is executed entirely on a remote computer or server, the remote computer can be connected to the user's computer via any type of network, including local area network (LAN) or wide area network (WAN), or can be connected to External computer (for example, using an internet service provider, via the internet).

Aspects of the present invention are described below with reference to flowchart illustrations and/or block diagrams of methods, apparatus (system) and computer program products according to embodiments of the present invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of the block diagrams and/or blocks in the block diagrams can be implemented by computer program instructions. The computer program instructions can be provided to a processor of a general purpose computer, a special purpose computer or other programmable data processing device to generate a machine for execution by the processor of the computer or other programmable data processing device The instructions are generated for implementation in the or the flowcharts And/or the components of the function/action specified in the block diagram.

The computer program instructions can also be stored in a computer readable medium that directs the computer, other programmable data processing device, or other device to function in a particular manner for storage on the computer readable medium. The instructions in the middle generate a production item that includes instructions for implementing the functions/actions specified in the flowcharts and/or block diagrams.

The computer program instructions can also be loaded onto a computer, other programmable data processing device or other device to cause a series of operational steps to be performed on the computer, other programmable device or other device to produce a computer implementation The program causes instructions executed on the computer or other programmable device to provide a program for implementing the functions/acts specified in the flowcharts and/or block diagrams.

Embodiments of the present invention can be provided to end users via a cloud computing infrastructure. Cloud computing generally refers to providing scalable computing resources as a service over a network. More formally, cloud computing can be defined as providing the ability to compute abstractions between computing resources and their underlying technology architecture (eg, servers, storage, networking), enabling one of the configurable computing resources Shared pools for convenient, on-demand network access, which can be quickly deployed and distributed with minimal administrative effort or service provider interaction. Therefore, cloud computing allows users to access virtual computing resources (eg, storage, data, applications, and even fully virtualized computing systems) in the "cloud" without regard to the underlying entities used to provide such computing resources. System (or the location of their systems).

Typically, cloud computing resources are provided to users based on usage-based billing, only for the computing resources actually used (eg, storage space consumed by the user or by the user to perform individualized virtualization systems) The amount) charges the user. The user can access any of the resources residing in the cloud at any time and from anywhere on the Internet. In the context of the present invention, the user has access to applications or related materials available in the cloud that are executed on the server or stored on the server. For example, an application can be executed on a server that implements a virtual switch in the cloud. Doing so allows the user to access this information from any computing system attached to a network connected to the cloud (eg, the Internet).

1 illustrates a system architecture including a virtual switch in accordance with an embodiment of the present invention. The first server 105 can include at least one processor 109 coupled to the memory 110. Processor 109 may represent one or more processors (eg, a microprocessor) or a multi-core processor. The memory 110 can represent a random access memory (RAM) device including the main memory of the server 105, and a supplemental level of memory, such as cache memory, non-volatile or backup memory (eg, , programmable memory or flash memory), read-only memory, and the like. Additionally, memory 110 can include a memory bank physically located in server 105 or on another computing device coupled to server 105.

The server 105 is operable under the control of the operating system 107 and executes various computer software applications, components, programs, objects, modules, and data structures (such as the virtual machine 111).

The server 105 can include a network adapter 115 (eg, a converged network Connector). Converged network adapters can include single root I/O virtualization (SR-IOV) adapters, such as Fast Peripheral Component Interconnect (PCIe) mating to support Converged Enhanced Ethernet (CEE) Device. Another embodiment of system 100 can include a multi-root I/O virtualization (MR-IOV) adapter. The network adapter 115 can further be used to implement the Fibre Channel over Ethernet (FCoE) protocol, the Ethernet RDMA, the Internet Small Computer System Interface (iSCSI), and the like. In general, the network adapter 115 transmits data using an Ethernet or PCI based communication method and can be coupled to one or more of the virtual machines 111. Additionally, the adapter can facilitate shared access between virtual machines 111. While the adapter 115 is shown as being included within the server 105, in other embodiments, the adapter can be a physically distinct device separate from the server 105.

In one embodiment, each network adapter 115 can include a polymeric adapter virtual bridge (not shown) that coordinates access to the virtual machine 111 by coordinating The data transfer between the adapters 115 is promoted. Each polymeric adapter virtual bridge can identify the flow of data within its domain (ie, the addressable space). The identified domain address can be directly routed without transmitting the data outside of the domain of the particular aggregated adapter virtual bridge.

Each network adapter 115 can include one or more Ethernet ports coupled to one of the bridge elements 120. Additionally, to facilitate PCIe communication, the server can have a PCI host bridge 117. The PCI host bridge will then be connected to the upstream PCI port 122 on the switch element in the virtual switch 180. The data is then routed via switch layer 130 to the correct downstream PCI port 123, which may be located in the same upstream PCI port 122 Or on different switch modules. This information can then be forwarded to the PCI device 150. As used herein, a "virtual switch" is used to describe the plurality of decentralized interconnects that appear to be a single switch for a device connected to one of a plurality of distributed hardware switch components. Hardware switch component.

Bridge component 120 can be configured to relay data frames throughout distributed virtual switch 180. For example, network adapter 115 and bridge component 120 can be connected using two 40 Gbit Ethernet connections or one 100 Gbit Ethernet connection. The bridge component 120 forwards the data frame transmitted by the network adapter 115 to the switching layer 130. The bridge component 120 can include a lookup table that stores address data for transmitting the received data frame. For example, bridge component 120 can compare the address data associated with the received data frame with the address data stored in the lookup table. Therefore, network adapter 115 does not need to know the network topology of distributed virtual switch 180.

In general, distributed virtual switch 180 includes a plurality of bridge elements 120 that can be located on a plurality of separate (but interconnected) hardware components. From the perspective of network adapter 115, virtual switch 180 functions as a single switch, although switch 180 may be comprised of multiple switches physically located on different components. In the event of a failure, the switches 180 are distributed to provide redundancy and expanded to interconnect more servers in a more cost effective manner relative to a single large switch.

Each of the bridge elements 120 can be coupled to one or more transport layer modules 125 that relay the received data frames Translated into the agreement used by the exchange layer 130. For example, the transport layer module 125 can translate the data received using the Ethernet or PCI communication method into a common format (ie, cell) transmitted via the switch layer 130 (ie, the cellular channel network). The package. Thus, the switch module containing switch 180 is compatible with at least two different communication protocols, such as Ethernet and PCI communication standards. That is, at least one switch module has the necessary logic to transfer different agreed payloads on the same switch fabric 130.

Although not shown in FIG. 1, in an embodiment, the switch layer 130 can include an area rack interconnect (which connects the bridge elements 120 located in the same chassis and rack) and connects to other chassis and racks. The connection of the bridge component 120.

After routing the cells, the switching layer 130 can communicate with the transport layer module 126, which translates the cells back to the data frame or message corresponding to the appropriate communication protocol. Portions of bridge component 120 facilitate communication with Ethernet 155, which provides access to a LAN or WAN (e.g., the Internet). In addition, the PCI data can be routed to a downstream PCI port 123 that is connected to the PCIe device 150. The PCIe device 150 can be a passive backplane interconnect (as an expansion card interface for add-on boards) or a common storage accessible by any of the servers connected to the switch 180.

The second server 106 can include a processor 109 coupled to the operating system 107 and a memory 110 including one or more virtual machines 111 similar to those found in the first server 105. The memory 110 of the server 106 also includes a hypervisor 113 having a virtual bridge 114. The hypervisor 113 manages materials shared between different virtual machines 111. Specifically, the virtual bridge The connector 114 allows direct communication between the connected virtual machines 111 without requiring the virtual machine 111 to use the bridge element 120 or the switching layer 130 to transmit data to other virtual machines communicatively coupled to the hypervisor 113. 111.

An input/output management controller (IOMC) 140 (ie, a dedicated processor) is coupled to at least one bridge element 120 that provides access to the switching layer 130 to the IOMC 140. One of the functions of the IOMC 140 is to receive commands from the administrator to configure different hardware components of the decentralized virtual switch 180. In an embodiment, such commands may be received from a separate switching network from the switching layer 130.

Although one IOMC 140 is shown, system 100 can include a plurality of IOMCs 140. In an embodiment, the IOMCs 140 may be configured in a hierarchical architecture such that an IOMC 140 is selected as the master while others are delegated as components (or slaves).

FIG. 2 illustrates a hardware hierarchy diagram of system 200 in accordance with an embodiment. Servers 210 and 212 can be physically located in the same chassis 205; however, chassis 205 can include any number of servers. The chassis 205 also includes a plurality of switch modules 250, 251 that include one or more sub-switches 254. In one embodiment, the switch modules 250, 251, 252 are hardware components (eg, PCB boards, FPGA boards, systems) that provide physical support and connectivity between the network adapter 115 and the bridge component 120. Single chip, etc.). In general, the switch modules 250, 251, 252 include hardware that connects the different chassis 205, 207 and servers 210, 212, 214 in the system 200.

The switch modules 250, 251, 252 (i.e., chassis interconnect elements) include one or more sub-switches 254 and IOMCs 255, 256, 257. Sub-switch 254 can include logical or physical groupings of bridge elements 120. Each bridge element 120 can be physically coupled to the servers 210, 212. For example, bridge component 120 can route data transmitted using Ethernet or PCI communication protocols to other bridge components 120 attached to switch fabric 130. However, in an embodiment, the bridge element 120 may not need to provide connectivity from the network adapter 115 to the switch layer 130 for PCI or PCIe communication.

Each switch module 250, 251, 252 includes IOMCs 255, 256, 257 for managing and configuring different hardware resources in system 200. In one embodiment, each of the IOMCs of each of the switch modules 250, 251, 252 can be responsible for configuring hardware resources on a particular switch module. However, because the switch modules are interconnected using the switch fabric 130, the IOMC on a switch module can manage the hardware resources on different switch modules.

The dotted lines in the chassis 205 define a midplane 220 between the servers 210, 212 and the switch modules 250, 251. That is, the midplane 220 includes a data path for transferring data between the network adapter 115 and the sub-switch 254.

Each bridge element 120 is connected to the exchange layer 130. Additionally, bridging component 120 can also be coupled to network adapter 115 or uplink. As used herein, the uplink of bridging component 120 provides services that extend the connectivity or capabilities of system 200. As shown in chassis 207, a bridging component 120 includes a connection to an Ethernet or PCI connector 260. For Ethernet communications, connector 260 can provide system 200 with access to a LAN or WAN (e.g., the Internet). Alternatively, the port connector 260 can connect the system to a PCIe expansion slot, such as the PCIe device 150. Device 150 can be an additional memory or memory, and each server 210, 212, 214 can be via switching layer 130 To access the extra storage or memory. Advantageously, system 200 provides access to switch fabric 130 having network devices that are compatible with at least two different communication methods.

As shown, the servers 210, 212, 214 can have a plurality of network adapters 115. This situation provides redundancy in the event that one of the adapters 115 is faulty. Additionally, each adapter 115 can be attached to a different switch module 250, 251, 252 via a midplane 220. As illustrated, one of the adapters 210 is communicatively coupled to the bridge component 120 located in the switch module 250 while the other adapter is coupled to the bridge component in the switch module 251. 120. If one of the switch modules 250, 251 is faulty, the server 210 can still access the switch layer 130 via other switch modules. The faulty switch module can then be replaced (e.g., hot swapped), which causes the IOMCs 255, 256, 257 and bridge component 120 to update the routing table and lookup table to include the hardware components on the new switch module.

Figure 3 illustrates a virtual switching layer in accordance with an embodiment of the present invention. Each of the sub-switches 254 of the systems 100 and 200 are connected to each other via a mesh connection schema using a switch layer 130 and a bridge component 120. That is, regardless of the sub-switch 254 used, the cells (i.e., data packets) can be routed to another other sub-switch 254 located on any of the other switch modules 250, 251, 252. This can be accomplished by directly connecting each sub-switch 254 - that is, each sub-switch 254 has a dedicated data path to each of the other sub-switches 254. Alternatively, the exchange layer 130 can use a spine-leaf architecture in which each sub-exchanger 254 (ie, a leaf node) is attached to at least one spinal node. Spinal node The cells received by the converter 254 are routed to the correct destination sub-switch 254. However, the invention is not limited to any particular technique for interconnecting sub-switches 254.

Network Fibre Channel network implementing VLAN

4 illustrates the use of a virtual area access network in the system of FIG. 2, in accordance with an embodiment of the present invention. Specifically, FIG. 4 simplifies FIG. 2 to include a server 210 and a sub-switch 254. Network adapter 115 is coupled to bridge element 120 via midplane 220. However, the midplane 220 is simplified to show only one of the physical connections that make up the LAN between the network adapter 115 and the bridge element 120. In an embodiment, each network adapter 115 will have one or more connections to the respective bridge elements 120.

The dashed lines define different VLAN channels 405A through 405C between the network adapter 115 and the particular bridge element 120. Although network adapter 115 may have only one or two physical Ethernet connections to bridge component 120 (eg, a 100 G/bit connection or two 40 G/bit connections), such connections may It is logically divided into VLANs including one or more VLAN channels for point-to-point communication. That is, each physical connection in the midplane 220 can be divided into thousands (if not millions) of different VLANs and their respective VLAN channels. In addition, as defined in IEEE 802.1QinQ, a VLAN channel can be an S channel or a C channel for an S-VLAN or a C-VLAN. Assigning packets to VLANs allows for finer control over the LAN. For example, an administrator can assign different security protocols to different VLANs or manage packets for each VLAN in different ways.

In FIG. 4, each network adapter 115 can be established to bridge component 120 One or more VLAN channels are provided as long as the adapter 115 is physically coupled to the bridge element 120 via the midplane 220. Thus, the uppermost network adapter 115 can have a VLAN channel to other bridge elements 120 than one of the bridge components shown. System 400 can assign as many VLAN channels as there are uniquely addressable VLANs. For IEEE 802.1Q, this is 4096 S-VLANs. The VLAN channels can be dispersed between the different physical connections of the midplane 220 in any manner. Note that each physical connection according to 802.1Q can have the same configured VLAN channel (or a subset of VLAN channels).

FIG. 5 illustrates a data frame for identifying network traffic based on a VLAN ID, in accordance with an embodiment of the present invention. In particular, data frame 500 illustrates one portion of a data frame that is compatible with IEEE 802.1QinQ. The data frame 500 is divided into two distinct parts: an external tag (S tag) 501 and an internal tag (C tag) 502. The external tag 501 is the same as the one defined in IEEE 802.1Q that first introduces the VLAN into the CEE network. However, in order to provide administrators with greater control over the LAN, IEEE 802.1QinQ introduces the concept of adding dual tags by adding internal tags 502. Thus, the embodiments disclosed herein can be used to address VLANs established based on IEEE 802.1Q or 802.1QinQ.

The tags 501, 502 include an Ethernet type/size 505, a priority code point (PCP) 510, a standard format indicator (CFI) 515, and an S-VLAN ID 520 or C-VLAN ID 525. The Ethernet type 505 is typically a 16-bit (or two-byte) field that indicates which protocol is encapsulated in the payload of the data packet. PCP 510 is a three-dimensional field, which refers to The priority or traffic class assigned to the S-VLAN or C-VLAN (ie, priority 1 to priority 8). CFI 515 is a 1-bit field that indicates whether the MAC address (stored elsewhere in data frame 500) is in a non-standard format.

For IEEE 802.1Q and 802.1QinQ, the S-VLAN ID 520 is 12 bits and specifies the S-VLAN to which the frame 500 belongs. The C-VLAN ID that is not defined in the IEEE 802.1Q standard is also 12 bits long and specifies the C-VLAN to which the frame 500 belongs. That is, IEEE 802.1QinQ extends IEEE 802.1Q such that each S-VLAN can have up to 4096 uniquely addressable C-VLANs, thereby providing more than 16 million addressable VLANs. However, the data frame defined only by IEEE 802.1Q will include portions 505A, 510A, 515A, and 520 but will not include portions 505B, 510B, 515B, and 525. In either case, the embodiments disclosed herein are not limited to IEEE 802.1Q and 802.1QinQ, but may be used with any relevant standard or derivative standard of such standards (eg, IEEE 802.1Qat or IEEE 802.lak) or current or It will be compatible with any other standard that produces VLANs in the future.

In general, data frame 500 is part of a larger data frame (i.e., data packet) that is transmitted from the source to the destination across the network. The entire data frame will include other partitions including, for example, preambles, source and destination addresses, payload, error correction codes, and the like.

Allocate bandwidth in a VLAN channel

6 illustrates an enhanced version of an ETS using a VLAN in the system of FIG. 1 in accordance with an embodiment of the present invention. Figure 6 illustrates the bridge component 120, memory Body 639 and 埠 630. Memory 639 can represent random access memory (RAM) device cache memory, non-volatile or backup memory (eg, programmable memory or flash memory), read-only memory, and the like. . Memory 639 includes an ETS scheduler 640 that divides the bandwidth associated with the physical connection between two computing devices (eg, between two bridge elements 120). In an embodiment, ETS scheduler 640 is configured to comply with IEEE 802.1Qaz.

The 埠 630 is used by the bridge component 120 to establish a connection to the computing device. For example, the port 630 can be a physical connector to which an Ethernet cable is attached. As shown, the UI 630 can be logically partitioned to represent a bandwidth allocation for a VLAN channel that uses the physical connection. For example, ETS scheduler 640 can be configured to allocate bandwidth based on the number of VLAN channels or VLANs in a Fibre Channel network. A certain percentage of the bandwidth available for 埠 630 can be guaranteed for each VLAN channel. If one of the VLAN channels is not currently using its allocated bandwidth, the ETS scheduler 640 can temporarily allow the network traffic associated with a different VLAN channel to use its bandwidth.

In addition, the bandwidth allocation can be assigned to one of the blocks or groups of VLAN channels. That is, VLAN channels 1 through 3 can be assigned 2% of the bandwidth. Moreover, in an embodiment, ETS scheduler 640 can allocate the bandwidth to a subset of the VLAN channels using 埠 630. Therefore, not every VLAN channel that uses 埠 630 must be allocated or guaranteed by the ETS scheduler 640.

In an embodiment, the ETS scheduler 640 can be located in memory elsewhere in the virtual switch 180.

7A-7B illustrate an ETS scheduler hierarchy architecture in accordance with an embodiment of the present invention. Instead of relying solely on allocating bandwidth to a single VLAN channel or a group of VLAN channels, the ETS scheduler can also assign bandwidth to the traffic classes within the VLAN. As shown by data frame 500 in FIG. 5, each frame assigned to a VLAN can also be assigned to a particular traffic class within the VLAN (i.e., the value represented by PCP 510). Each data frame can be assigned to a VLAN and a particular traffic class based on the VLAN ID of the VLAN (ie, S-VLAN ID 520 and/or C-VLAN ID 525).

The memory 739 can include an ETS scheduler hierarchy consisting of an ETS VLAN scheduler 741 and an ETS priority scheduler 742. The ETS VLAN Scheduler 741 can perform functions similar to the ETS Scheduler 640 of FIG. That is, the ETS VLAN scheduler 741 can allocate the bandwidth to one or more VLAN channels using the egress port 730. In addition to allocating bandwidth between VLAN channels, the ETS priority scheduler 742 of the ETS scheduler hierarchy can further divide the bandwidth between traffic classes. That is, the ETS priority scheduler 742 can further divide the bandwidth allocated to a specific VLAN channel between different traffic classes (ie, priority levels 1 to 8) in the VLAN channel.

Figure 7B illustrates an example for allocating bandwidths assigned to VLAN channels or VLANs based on traffic classes. As shown, the ETS VLAN Scheduler 741 allocates 15% of the bandwidth associated with the UI 730 to the network traffic assigned to VLAN Channel 1. In addition, the ETS priority scheduler 742 further divides the 15% allocation between different traffic classes (i.e., priority levels 1 through 8). Therefore, priority 8 receives 6% of the total bandwidth of 埠730, while priority 1 receives only 1% of the total bandwidth. Correspondingly, the priority assigned to VLAN channel 1 Level 8 network traffic guarantees at least 6% of the bandwidth and 1% to priority 1. In this way, the bandwidth of 埠730 can be divided into any desired distribution.

In addition, the ETS scheduler hierarchy can be used to allocate bandwidth when multiple VLAN channels are grouped together. For example, if the ETS VLAN scheduler 741 allocates 15% of the bandwidth of the bandwidth to VLAN channels 1 through 3, the ETS priority scheduler 742 can assign a portion of the bandwidth to the VLAN channels. Each of the associated traffic categories. For example, priority 1 of VLAN channel 1 and priority 5 of VLAN channel 2 can receive all bandwidths, while other priorities are not allocated bandwidth. Alternatively, the ETS priority scheduler 742 can treat each of the traffic classes of the respective VLAN channels equally. In this case, the bandwidth allocation will be similar to the bandwidth allocation shown in Figure 7B, with the exception that the allocation is applied to multiple VLAN channels. That is, the priority 1 (as a group) of each of the VLAN channels will be allocated 1% of the total bandwidth, the priority 5 of each of the VLAN channels will be allocated a total of 2%, and so on, Until the 15% allocation for multiple VLAN channels has been distributed. However, similar to VLANs, not all traffic classes within a VLAN must be allocated bandwidth, even if the VLAN itself is allocated bandwidth.

Although shown as a separate component, in one embodiment, ETS VLAN scheduler 741 and ETS priority scheduler 742 can be integrated into a single component.

Figure 8 is a simplified representation of the system illustrated in Figure 2. Figure 8 illustrates the connection between a server and a network device (i.e., bridge component 120), between network devices, between a network device and other WANs or LANs, and between network devices and other computing devices. Connection.

As shown, Figure 8 includes a server 805 having a virtual machine 111, which is virtual The projector 111 can have two different applications 810, 815 being executed. The applications 810, 815 are each associated with a separate VLAN (or more specifically, an S-channel) that routes data packets from the application 810, 815 to at least one of the bridge elements 120 of the virtual switch 180. For the sake of clarity, the physical connection between the server 805 and the bridge element 120 is omitted. Additionally, VLAN channels 405D and 405E can be associated with the same connection or different physical connections.

The virtual switch 180 includes a plurality of bridge elements 120A-120C that include an inlet port 825, outlet ports 830, 835, and 838, and an ETS scheduler 840. Although not shown, the server 805 can also include an ETS scheduler and an exit port that is logically partitioned to allocate bandwidth based on VLANs in the network.

Bridge component 120A receives data packets (i.e., data frames) from two applications 810, 815 that are assigned to respective VLAN channels. Using a queue, bridge component 120A can use a mesh network in virtual switch 180 to forward packets to different network devices in virtual switch 180 via egress 埠 830. The data path 870 associated with the application 815 transmits the data frame assigned to the VLAN channel 405E. The ETS scheduler 840 of the bridge component 120A can distribute the bandwidth to the network traffic and the data traffic flowing across the VLAN channels 405D and 405E in the direction toward the server 805.

After the ETS scheduler 840 establishes a bandwidth allocation for the UI 830, when the bridge component 120A receives the data frame, it evaluates the VLAN ID to determine which VLAN it belongs to. When the packet is subsequently transferred, the bridge element 120A Know how much bandwidth is available for this VLAN. For example, if the ETS scheduler 840 allocates 4% of the bandwidth to the VLAN, then at least as many data frames are guaranteed to be transmitted when all data frames having the corresponding VLAN ID are forwarded across the corresponding VLAN channel. bandwidth.

If the ETS scheduler 840 includes an ETS scheduler hierarchy, the scheduler 840 can further divide the allocated bandwidth based on the traffic class within the VLAN containing the VLAN channel 405E. The ETS scheduler 840 can execute the same procedure (ie, data path 875) for the VLAN channel 405D for the application 810. However, the priority code and VLAN ID in the received data frame are used to ensure that the correct amount of bandwidth is used when forwarding the packet. In this manner, ETS scheduler 840 can ensure that the VLAN and traffic classes of data paths 870 and 875 have a minimum percentage of available bandwidth of 埠 830.

However, the ETS scheduler 840 for the bridge elements 120B and 120C can assign different percentages of bandwidth to the data paths 870 and 875. That is, the bandwidth allocation for the network traffic to the data path 870 flowing through the bridge component 120A can be, for example, greater than or less than the bandwidth allocation of the network traffic given to the data path 870 flowing through the bridge component 120C. . Alternatively, the ETS scheduler 840 in the bridge component 120C may not distribute the bandwidth to the data frame of the data path 870 or the data frame of the data path 870. In general, the system administrator can configure the ETS scheduler 840 of the bridge elements 120A-120C to meet the needs of a particular bridge element 120. For example, bridge component 120B can forward data traffic to different WANs, while bridge component 120C can forward data traffic to computing devices such as another server or storage device connected to virtual switch 180. . These different configurations can make The administrator has differently assigned bandwidths for different bridge elements 120.

An ETS as defined in IEEE 802.1Qaz does not provide a way to allocate the bandwidth of a physical connection with any subtlety that is greater than the granularity of the traffic class. As mentioned previously, IEEE 802.1Q allows network administrators to establish up to 4096 different VLANs, while IEEE 802.1QinQ provides more than 16 million VLANs. The ability to allocate bandwidth based on VLANs rather than traffic classes increases the ability of network administrators to control how bandwidth is allocated. The ETS scheduler 840 disclosed herein can allocate different portions of the bandwidth to the application 810 (i.e., VLAN channel 405D) and the application 815 (i.e., VLAN channel 405E). Instead of having to assign VLAN channels 405D, 405E to different traffic classes to differently allocate bandwidth between applications, ETS scheduler 840 can be configured to be based on members of the received data frame in a particular VLAN. Eligibility assigns the bandwidth to the received data frame.

The exit ports 830, 835, 838 can be divided into thousands or millions of different allocations based on VLAN IDs (eg, S-VLANs and C-VLANs). Currently, ETS provides bandwidth allocation in 1% increments (ie, up to 100 VLANs can be allocated 1% of the bandwidth); however, the embodiments disclosed herein can provide much greater precision. To describe the future bandwidth allocation structure of the allocated bandwidth. The future structure description may permit the ETS scheduler 840 to divide the bandwidth of the physical connection into any different combination between one and sixteen million different VLANs that may currently use IEEE 802.1QinQ. This number can be further extended if an ETS scheduler hierarchy based on eight traffic classes is used to subdivide the bandwidth allocated to the VLAN.

in conclusion

IEEE 802.1Q and ETS provide only eight different traffic classes that can be used to control the bandwidth in a particular physical connection (or link). Instead of relying solely on these eight traffic classes to manage bandwidth, the embodiments discussed herein disclose the use of an enhanced ETS scheduler that allows the network device to set the bandwidth for individual VLANs. Bandwidth allocation based on VLAN ID allows network devices to allocate bandwidth to millions of unique VLANs. Therefore, this technology can increase the fine-grained control of the network's Fibre Channel network and its performance.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products in accordance with various embodiments of the present invention. In this regard, each block of the flowchart or block diagram can represent a module, a segment or a portion of a code, which comprises one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks may occur in the order noted in the figures. For example, two blocks of consecutive presentations may be executed substantially concurrently, or the blocks may sometimes be performed in the reverse order, depending on the functionality involved. It should also be noted that each block and block diagram and/or process of the block diagrams and/or flowchart illustrations can be implemented by a dedicated hardware-based system or a combination of dedicated hardware and computer instructions that perform the specified function or action. The combination of blocks in the illustration.

While the foregoing is directed to the embodiments of the present invention, the subject matter of the present invention can be devised by the scope of the following claims.

100‧‧‧ system

105‧‧‧First server

106‧‧‧Second server

107‧‧‧Operating system

109‧‧‧Processor

110‧‧‧ memory

111‧‧‧Virtual Machine

113‧‧‧Super Manager

114‧‧‧Virtual Bridge

115‧‧‧Network adapter

117‧‧‧PCI (Peripheral Component Interconnect) Host Bridge

120‧‧‧Bridge components/bridge components

120A‧‧‧Bridge components

120B‧‧‧Bridge components

120C‧‧‧Bridge components

122‧‧‧Upstream PCI (Peripheral Component Interconnect)埠

123‧‧‧Downstream PCI (Peripheral Component Interconnect)埠

125‧‧‧Transport layer module

126‧‧‧Transport layer module

130‧‧‧ exchange layer

140‧‧‧Input/Output Management Controller (IOMC)

150‧‧‧PCI (peripheral component interconnect) devices/PCIe (fast peripheral component interconnect) devices

155‧‧‧Ethnet

180‧‧‧Virtual Switch

200‧‧‧ system

205‧‧‧Chassis

207‧‧‧Chassis

210‧‧‧Server

212‧‧‧Server

214‧‧‧ server

220‧‧‧ midplane

250‧‧‧Switch Module

251‧‧‧Switch Module

252‧‧‧Switch Module

254‧‧‧Sub-switch

255‧‧‧IOMC (input/output management controller)

256‧‧‧IOMC (input/output management controller)

257‧‧‧IOMC (Input/Output Management Controller)

260‧‧‧Ethernet or PCI (peripheral component interconnect) connector/埠 connector

400‧‧‧ system

405A‧‧‧VLAN (Virtual Area Network) channel

405B‧‧‧VLAN (Virtual Area Network) channel

405C‧‧‧VLAN (Virtual Area Network) channel

405D‧‧‧VLAN (Virtual Area Network) channel

405E‧‧‧VLAN (Virtual Area Network) channel

500‧‧‧Information frame/frame

501‧‧‧External label (S label) / external label

502‧‧‧Internal label (C label) / internal label

Section 505A‧‧‧

Section 505B‧‧‧

Section 510A‧‧‧

Section 510B‧‧‧

Section 515A‧‧‧

Section 515B‧‧‧

520‧‧‧S-VLAN (service-virtual area network) ID (identifier) / part

525‧‧‧C-VLAN (Customer-Virtual Area Network) ID (Identifier) / Part

630‧‧‧埠

639‧‧‧ memory

640‧‧‧ETS (Enhanced Transmission Selection) Scheduler

730‧‧‧Export

739‧‧‧ memory

741‧‧‧ETS (Enhanced Transmission Selection) VLAN (Virtual Area Network) Scheduler

742‧‧‧ETS (Enhanced Transmission Selection) Priority Scheduler

805‧‧‧Server

810‧‧‧Application

815‧‧‧Application

825‧‧‧Entry埠

830‧‧‧ Export 埠

835‧‧‧ Export 埠

838‧‧‧Export

840‧‧‧ETS (Enhanced Transmission Selection) Scheduler

870‧‧‧ data path

875‧‧‧ data path

1 illustrates a system architecture including a decentralized virtual switch in accordance with an embodiment of the present invention.

2 illustrates a hardware representation of a system implementing a virtual switch in accordance with an embodiment of the present invention.

Figure 3 illustrates a virtual switch in accordance with an embodiment of the present invention.

4 illustrates the use of a virtual area access network in the system of FIG. 1 in accordance with an embodiment of the present invention.

FIG. 5 illustrates a data frame for identifying network traffic based on a virtual area network ID, in accordance with an embodiment of the present invention.

6 illustrates an enhanced version of the enhanced transmission selection used in the system of FIG. 1 in accordance with an embodiment of the present invention.

7A-7B illustrate an enhanced version of enhanced transmission selection using a virtual area network, in accordance with an embodiment of the present invention.

8 illustrates an enhanced version of enhanced transmission selection using a virtual local area network in the system of FIG. 2, in accordance with an embodiment of the present invention.

100‧‧‧ system

105‧‧‧First server

106‧‧‧Second server

107‧‧‧Operating system

109‧‧‧Processor

110‧‧‧ memory

111‧‧‧Virtual Machine

113‧‧‧Super Manager

114‧‧‧Virtual Bridge

115‧‧‧Network adapter

117‧‧‧PCI (Peripheral Component Interconnect) Host Bridge

120‧‧‧Bridge components/bridge components

122‧‧‧Upstream PCI (Peripheral Component Interconnect)埠

123‧‧‧Downstream PCI (Peripheral Component Interconnect)埠

125‧‧‧Transport layer module

126‧‧‧Transport layer module

130‧‧‧ exchange layer

140‧‧‧Input/Output Management Controller (IOMC)

150‧‧‧PCI (peripheral component interconnect) devices/PCIe (fast peripheral component interconnect) devices

155‧‧‧Ethnet

180‧‧‧Virtual Switch

Claims (21)

A method for transmission, comprising: establishing a physical connection connecting one of two computing devices, wherein network traffic flowing through the physical connection is logically divided into a plurality of virtual area network (VLAN) channels; Network traffic of at least one of the plurality of VLAN channels, a bandwidth of the physical connection is allocated between at least two of the plurality of VLAN channels; and the plurality of traffic classes are further allocated to The bandwidth of one of the plurality of VLAN channels. The method of claim 1, wherein the plurality of VLAN channels and the plurality of traffic classes are configured based on at least one of: IEEE 802.1Q and one of IEEE 802.1Q derivatives. The method of claim 1, further comprising: determining whether the network traffic assigned to the one of the plurality of VLAN channels is using the allocated bandwidth, if assigned to the plurality of VLAN channels If the network traffic of the one does not use the allocated bandwidth, assign at least a portion of the bandwidth assigned to the one of the plurality of VLAN channels to a different VLAN channel; The portion of the bandwidth is reassigned from the different VLAN channels to the one of the plurality of VLAN channels. The method of claim 1, wherein the VLAN channel is associated with a service VLAN (S-VLAN). The method of claim 1, wherein the VLAN channel is associated with a customer VLAN (C-VLAN) assigned to one of the S-VLANs. The method of claim 5, wherein the S-VLAN and the C-VLAN are configured based on at least one of: IEEE 802.1QinQ and one of IEEE 802.1QinQ derivatives. The method of claim 1, wherein only a subset of the plurality of VLAN channels to which the entity is connected is allocated a bandwidth. A computer program product for transmission, comprising: a computer readable storage medium having computer readable code embodied therein, the computer readable code comprising computer readable code configured to perform the following operations Establishing a physical connection connecting one of the two computing devices, wherein the network traffic flowing through the physical connection is logically divided into a plurality of virtual local area network (VLAN) channels; based on flowing through at least one of the plurality of VLAN channels Network traffic, a bandwidth of the physical connection is allocated between at least two of the plurality of VLAN channels; and one of the plurality of VLAN channels is further divided among the plurality of traffic classes The bandwidth. The computer program product of claim 8, wherein the plurality of VLAN channels and the plurality of traffic classes are configured based on at least one of: IEEE 802.1Q and one of IEEE 802.1Q derivatives. The computer program product of claim 8 further comprising computer readable code configured to: Determining whether the network traffic assigned to the one of the plurality of VLAN channels is using the allocated bandwidth, and if the network traffic assigned to the one of the plurality of VLAN channels is not Using the allocated bandwidth, assigning at least a portion of the bandwidth assigned to the one of the plurality of VLAN channels to a different VLAN channel; and re-establishing the portion of the bandwidth from the different VLAN channel Assigned to one of the plurality of VLAN channels. The computer program product of claim 8, wherein the VLAN channel is associated with a service VLAN (S-VLAN). The computer program product of claim 8, wherein the VLAN channel is associated with a customer VLAN (C-VLAN) assigned to one of the S-VLANs. The computer program product of claim 12, wherein the S-VLAN and the C-VLAN are configured based on at least one of: IEEE 802.1QinQ and one of IEEE 802.1QinQ derivatives. The computer program product of claim 8, wherein only a subset of the plurality of VLAN channels to which the entity is connected is allocated a bandwidth. A network device comprising: a port configured to be coupled to a computing device via a physical connection, wherein network traffic flowing through the physical connection is logically divided into a plurality of virtual area networks (VLANs) a channel; and a bandwidth scheduler configured to distribute the network between at least two of the plurality of VLAN channels based on network traffic flowing through at least one of the plurality of VLAN channels One bandwidth and in multiple traffic categories The bandwidth allocated to one of the plurality of VLAN channels is further divided. The network device of claim 15, wherein the plurality of VLAN channels and the plurality of traffic classes are configured based on at least one of: IEEE 802.1Q and one of IEEE 802.1Q derivatives. The network device of claim 15, wherein the bandwidth scheduler is configured to: determine whether the network traffic assigned to the one of the plurality of VLAN channels is using the allocated bandwidth, If the network traffic assigned to the one of the plurality of VLAN channels does not use the allocated bandwidth, then assigning at least a portion of the bandwidth to the one of the plurality of VLAN channels Allocating to a different VLAN channel; and reallocating the portion of the bandwidth from the different VLAN channel to the one of the plurality of VLAN channels. A network device as claimed in claim 15, wherein the VLAN channel is associated with a serving VLAN (S-VLAN). A network device as claimed in claim 15, wherein the VLAN channel is associated with a customer VLAN (C-VLAN) assigned to one of the S-VLANs. The network device of claim 19, wherein the S-VLAN and the C-VLAN are configured based on at least one of: IEEE 802.1QinQ and one of IEEE 802.1QinQ derivatives. A network device as claimed in claim 15, wherein only a subset of the plurality of VLAN channels to which the entity is connected is allocated a bandwidth.